Evaluating the effectiveness of therapeutic horse riding in changing health outcomes for children experiencing disability: a critical realist approach

Background: Therapeutic horse riding (THR) is a complex intervention using horses and horse-related activities to positively influence health in people who experience disability. To date, no evaluation of THR effectiveness has been undertaken within New Zealand, and limited evaluations have been undertaken internationally. Evaluating the effectiveness of THR interventions is made more complex by a lack of clarity about ingredients, mechanisms of treatment effect, and intended treatment targets of THR. \n\nAim: To evaluate the effectiveness of a THR intervention – what works for which riders, in what contexts, to what extent and how.\n\nMethodology: Within a critical realist framework, three phases of research were undertaken using a mix of methods, culminating in an analytical synthesis in which findings from all three phases were integrated into one overall evaluative account.\n\nMethods: In the first phase both directed-content and thematic analysis were used to explore what is delivered to riders, the context in which THR is provided, and outcomes valued by various stakeholders. Issues related to diversity within key components and treatment targets of THR were also addressed. Data for this phase of research was collected from multiple sources, including from interviews and focus groups with stakeholders; from existing organisational documentation and training materials; from goal documentation forms; and from testimonials and feedback surveys completed by riders and caregivers. In the second phase, grounded theory methods were used to develop a model for understanding mechanisms of change in health outcomes for child riders. Data was collected from interviews and group conversations with riders and caregivers in their homes, and from field notes written in response to observation of THR sessions. Photographs taken of riders engaging in various THR session tasks were used to supplement data collection within interviews with riders immediately after their THR session. Phases 1 and 2 contributed to the determination of health outcomes measured in the third phase of research. In this phase, a randomised, multiple-baseline single-case experimental design, across 12 participants, was used to quantitatively evaluate in which riders, and to what extent, changes in balance, functional performance, social responsiveness, health-related quality of life and participation outcomes occurred as a result of a 20-week THR intervention. Analysis of data included analysis of individual participant responses (visual analysis, descriptive nonparametric approaches, and analyses considering baseline stability, evaluated using Mean Phase Difference and Slope and Level Change procedures) and between participant responses (Modified Brinley plots, estimation of effect sizes). When synthesising various analytical findings, responses that had a clear intervention effect and demonstrated meaningful clinical change were considered. Findings from all three phases were synthesised to provide an integrated evaluation of THR intervention effectiveness. Using context-mechanism-outcome configurations as a framework for synthesising findings, specific attention was paid to contextual factors and mechanisms of change which impacted on health outcome change for riders. \n\nResults: The first phase of research revealed that while stakeholders perceived THR as having a wide range of health outcome benefits, caregivers and riders prioritised THR as an activity that allowed riders to participate meaningfully in valued leisure roles. From Phase 2 a model was developed proposing that changes in health outcomes occurred in response to rider’s new experiences of learning to move, succeed, connect and adapt within the THR therapeutic landscape, thereby contributing to their developing self-concept. An inconsistent intervention response to THR was observed across participants in the SCED study. Carer-rated participation outcomes (i.e., functional performance in a range of settings other than THR sessions measured using the Canadian Occupation Performance Measure [COPM]) demonstrated the most consistent positive between phase difference (performance score ES = 1.23; satisfaction score ES = 1.11). However, when taking baseline data stability into account, these improvements only reached clinical significance for two participants. \nSynthesised results from all phases of research suggest that THR is an intervention that may contribute to improved participation outcomes for some children experiencing disability, regardless of their diagnosis. Findings indicate that THR is an accessible activity, allowing meaningful participation for children experiencing disability. Riders experience the THR therapeutic landscape as a context that promotes their capacities and strengths, rather than their deficits and difficulties. THR also provides opportunities for a broad range of learning experiences, with the child as an active agent within the intervention contributing to a rider’s enlarged self-concept regarding how they move, what they perceive themselves as being good at, who they connect with, and how well they cope with change.\n\nConclusion: Congruent with the critical realist framework adopted for this thesis, this work evaluated the effectiveness of THR by conceptualising a deeper understanding of relationships between contextual factors, mechanisms of effect and changes in health outcomes. A portable explanatory theory is proposed regarding how THR might exert its effects. The results from this thesis can be used to guide future research and THR provider training, rider outcome assessment and ongoing programme evaluation. THR providers may also find this thesis useful for critically appraising the way THR is delivered and for evaluating whether their current service approach to THR delivery matches the priorities of riders and caregivers.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.